© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-2B
    5. 21-GEMMASCOPE-TRANSCODER-16K
    7. 15503
    Prev
    Next
    INDEX
    Explanations

    seemingly unrelated terms, possibly including names, related to academic writing, and/or sexual content.

    oai_token-act-pair · gemini-2.0-flash

    words starting with 'm'

    np_token-act-pair-logits · gpt-4o-mini

    Mar

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_21/width_16k/average_l0_13
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.21.ln2.hook_normalized
    Architecture
    jumprelu_transcoder
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
     myſelf
    -1.41
     itſelf
    -1.29
     purpoſe
    -1.21
     Majefty
    -1.16
    AndEndTag
    -1.15
     poffe
    -1.14
     preſent
    -1.12
     poffible
    -1.10
     ſtate
    -1.10
     themſelves
    -1.10
    POSITIVE LOGITS
     th
    0.52
     có
    0.51
     tri
    0.50
     m
    0.50
    ko
    0.47
     konto
    0.47
    m
    0.47
    tri
    0.47
    maz
    0.46
     tan
    0.45
    Activations Density 0.495%

    No Known Activations